lukestein’s avatarlukestein’s Twitter Archive—№ 17,770

  1. …in reply to @mobariz_ahmad
    @mobariz_ahmad @tryvoiceink The two tricks are - voiceink (and many competitor tools) include text from the current onscreen window in the AI prompt as context, so the AI can find variable names and spelling - a prompt that basically says “turn the raw voice transcription into stata.” Here’s mine:
    oh my god twitter doesn’t include alt text from images in their API
    On twitter.com Twitter logo lukestein’s avatar ❤️ 3 Favorites Mood 0
    1. …in reply to @lukestein
      @mobariz_ahmad @tryvoiceink (Actually that’s two versions of the prompt. The first one was my handwritten one, and the second below was refined based on advice from Gemini.)
      Permalink On twitter.com Twitter logo lukestein’s avatar ❤️ 1 Favorite Mood +1 🙂
      1. …in reply to @lukestein
        @mobariz_ahmad @tryvoiceink More useful for anyone who wants to copy (or improve! please! tell me how!): gist.github.com/lukestein/c80a3feb018195f0cce2aeb21b2128a5#file-stata-md
        Permalink On twitter.com Twitter logo lukestein’s avatar ❤️ 2 Favorites Mood +5 🙂